Difference between revisions of "How to use Combo"

Revision as of 17:37, 25 March 2014

To start to use combo - our group cluster, you need to know something about it. This guide is intended for users with no prior experience with Linux/ Command Line Interface (CLI) or Cluster Computing. Refer to the User Guide for detailed information on using the Combo.

Introduction

The basic building block of a Linux cluster is a node. The two primary types of nodes on a cluster are:

• Compute nodes - these are the majority of nodes on a cluster and are the nodes that run user jobs
• Front-end or Head node(s) - one or more on a cluster and where you login from your local computer (laptop/desktop)

User access to the compute nodes is only available via a batch job. This is typically a sequence of commands listed in a file called a batch script that can be executed without the intervention of the user. The batch script is a plain text file that starts with directives that describe the requirements of the job such as number of nodes and wall clock time, followed by the user's commands to execute specific tasks or run a specific code. An interactive batch job provides a way to get interactive access to a compute node via a batch job. This is useful for testing and debugging purposes.

From the front-end, the batch job is submitted to a queue to be scheduled to run when the compute node resources requested are available. Users can also check status, monitor progress, or delete batch jobs on the head nodes.

IMPORTANT: The front-end on the Combo are a shared resource for all users of the system and their computational use should be limited to editing, compiling, and for short non-intensive runs. The administrators may terminate user processes on the head nodes that impact the system without warning. Users should do all production work by submitting jobs to the batch system.

How to Log in Combo

Use PuTTY on Windows or Terminal on Linux, login using your account and password. You need to use some specific IP address to access combo.

Only secured (ssh) connections to combo are supported. All insecure methods of connection (rlogin, rsh, telnet) are disabled. You will log on to the master node and you should be in your home directory which is also accessible by compute nodes.

If your local computer system does not support SSH, please install the SSH software [ SSH or Open SSH] before connecting to combo. For Microsoft Windows system, a free client called PuTTY is available here: http://www.chiark.greenend.org.uk/~sgtatham/putty/

First thing you need to do after first login

Change your password, in command line, type

$ passwd

to change your password. A good password will have a mix of lower- and upper-case characters, numbers, and punctuation marks, and should be at least 6 characters long. Some people like to put a punctuation mark/number in the middle of a word, e.g., Ha&pp6Y.

Second thing you need to do

You need to change ~/.bashrc file so that you can use the apps in shared folder easily, type

$ cp /share/apps/useful/setup/_.bashrc ~/.bashrc
$ source ~/.bashrc

to replace bashrc with prepared file. Then use this code to sync .bashrc to all nodes: (This step is optional, if you find the nodes does not apply same bashrc, then do it)

for i in compute-0-0 compute-0-2 compute-0-3 compute-0-4 compute-0-5 compute-0-6 compute-0-7 compute-0-8 compute-0-9 compute-0-10 compute-0-11 compute-0-12 compute-0-13 compute-0-14 compute-0-15 compute-0-16 ; do scp ~/.bashrc ${i}:~/ ; done
for i in compute-0-0 compute-0-2 compute-0-3 compute-0-4 compute-0-5 compute-0-6 compute-0-7 compute-0-8 compute-0-9 compute-0-10 compute-0-11 compute-0-12 compute-0-13 compute-0-14 compute-0-15 compute-0-16 ; do ssh ${i} source ~/.bashrc ; done

Backup your things

We should emphasize the different levels of importance of computer data. Most of the data on combo are generated by programs, and could be regenerated, in case of necessity. The source code, on the other hand, are very precious to the developers, and could not easily be regenerated. It is your responsibility to back up critical data! Please maintain your own copy of important data stored in the cluster. Users are highly recommended to do everything possible for ensuring those critical material (programs etc.) never be lost. For example, the users should keep multiple copies of these important files on different computer systems, including users' local system. The following example illustrates a directory move from combo:

• Create a copy of the local directory with tar (the time to do this depends upon the sizes and number of files, etc.):

$ tar -cvf name-of-your-file.tar directory_to_be_transfer

• Compress tar file with gzip (this step may not be necessary if your file is small):

$ gzip name-of-your-file.tar

This creates a compressed tar file with the name name-of-your-file.tar.gz

• Now put the tarball somewhere safe. Maybe a local driver or another server.
• Anytime you may move the tarball back to combo and untar:

$ tar -zxvf name-of-your-file.tar